paper-with-me

Papers

Conditional Generative Data-free Knowledge Distillation

2021-12-31 · Xinyi Yu, Ling Yan, Yang Yang, Libo Zhou, Linlin Ou

Knowledge distillation has made remarkable achievements in model compression. However, most existing methods require the original training data, which is usually unavailable due to privacy and security issues. In this paper, we propose a conditional generative data-free knowledge distillation (CGDD) framework for training lightweight networks without any training data. This method realizes efficient knowledge distillation based on conditional image generation. Specifically, we treat the preset labels as ground truth to train a conditional generator in a semi-supervised manner. The trained generator can produce specified classes of training images. For training the student network, we force it to extract the knowledge hidden in teacher feature maps, which provide crucial cues for the learning process. Moreover, an adversarial training framework for promoting distillation performance is constructed by designing several loss functions. This framework helps the student model to explore larger data space. To demonstrate the effectiveness of the proposed method, we conduct extensive experiments on different datasets. Compared with other data-free works, our work obtains state-of-the-art results on CIFAR100, Caltech101, and different versions of ImageNet datasets. The codes will be released.

📄 PDF Abstract BibTeX arXiv:2112.15358

Code (0)

등록된 구현이 없습니다.

Tasks

Conditional Image GenerationData-free Knowledge DistillationImage GenerationKnowledge DistillationModel Compression

Methods 이 논문이 사용한 방법론

Knowledge Distillation A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions.…

Similar Papers 제목 키워드 기반

Conditional Pseudo-Supervised Contrast for Data-Free Knowledge Distillation

2025-10-03 · Renrong Shao, Wei Zhang, Jun wang arxiv

Data-free knowledge distillation~(DFKD) is an effective manner to solve model compression and transmission restrictions while retaining privacy protection, which has attracted extensive attention in recent years. Current…

Data-free Knowledge DistillationContrastive LearningModel Compression

A Unified Knowledge Distillation Framework for Deep Directed Graphical Models

2021-09-29 · CVPR 2023 1 · Yizhuo Chen, Kaizhao Liang, Zhe Zeng, Yifei Yang 외

Knowledge distillation (KD) is a technique that transfers the knowledge from a large teacher network to a small student network. It has been widely applied to many different tasks, such as model compression and federate…

Continual LearningFederated LearningKnowledge DistillationModel Compression

Large-Scale Generative Data-Free Distillation

2020-12-10 · Liangchen Luo, Mark Sandler, Zi Lin, Andrey Zhmoginov 외

Knowledge distillation is one of the most popular and effective techniques for knowledge transfer, model compression and semi-supervised learning. Most existing distillation approaches require the access to original or a…

Knowledge DistillationModel CompressionTransfer Learning

FedDTG:Federated Data-Free Knowledge Distillation via Three-Player Generative Adversarial Networks

2022-01-10 · Lingzhi Gao, Zhenyuan Zhang, Chao Wu

While existing federated learning approaches primarily focus on aggregating local models to construct a global model, in realistic settings, some clients may be reluctant to share their private models due to the inclusio…

Data-free Knowledge DistillationFederated LearningGenerative Adversarial NetworkKnowledge Distillation

Random Conditioning for Diffusion Model Compression with Distillation

2025-01-01 · CVPR 2025 1 · Dohyun Kim, Sehwan Park, Geonhee Han, Seung Wook Kim 외

Diffusion models generate high-quality images through progressive denoising but are computationally intensive due to large model sizes and repeated sampling. Knowledge distillation--transferring knowledge from a comp…

DenoisingKnowledge DistillationmodelModel Compression